Generative AI and Assessments: An Exploratory Study in English Language Education

Abstract

The emerging technology of AI is becoming popular in the educational sector, as it offers new ways to design, deliver, analyze, and evaluate learning tasks. So, this exploratory research design, under the topic Generative AI and Assessments: An Exploratory Study in English Language Education, investigates the opportunities and challenges brought by generative artificial intelligence (AI) in English language education, focusing on its impact on pedagogy and assessment. For this, I have applied van Dijk’s Digital Divide Theory. This study reflects the views of six participants: three English language educators and three learners, all located within the Kathmandu Valley, as gathered through semi-structured interviews. To fulfill the purpose of how generative AI contributes to language learning, introduces pedagogical, ethical and assessment-related challenges within the context of the National Examination Board (NEB) assessment system in Nepal. The findings show that generative AI facilitates personalized learning, improves grammatical accuracy, expands vocabulary, and enhances writing fluency through automated guidance, support and access to authentic language input. The study had its drawbacks; it also posed significant challenges, such as overreliance on AI-generated responses, a lack of critical and mechanical engagement, and misalignment with assessment standards that value learners’ originality and contextual understanding. The results of this study underscore the indispensable role of human educators in bridging the gap between AI outputs and academic expectations in language learning preparedness and assessment practice. In conclusion, the study suggested that in this rapidly evolving tech-savvy era, students have to be taught ethical and critical AI literacy in the language classroom to support fair, creative, contextually relevant learning and assessment practices.